linux-mm.kvack.org archive mirror
 help / color / mirror / Atom feed
From: Andrea Arcangeli <andrea@cpushare.com>
To: linux-mm@kvack.org
Cc: Andrew Morton <akpm@linux-foundation.org>,
	David Rientjes <rientjes@google.com>
Subject: [PATCH 03 of 11] prevent oom deadlocks during read/write operations
Date: Thu, 03 Jan 2008 03:09:09 +0100	[thread overview]
Message-ID: <71f1d848763c80f336f7.1199326149@v2.random> (raw)
In-Reply-To: <patchbomb.1199326146@v2.random>

# HG changeset patch
# User Andrea Arcangeli <andrea@suse.de>
# Date 1199324664 -3600
# Node ID 71f1d848763c80f336f7f9f23dcbe8f7e43c82aa
# Parent  3ec0754b24738f5658baf2a1ebacc7d3f2c17283
prevent oom deadlocks during read/write operations

We need to react to SIGKILL during read/write with huge buffers or it
becomes too easy to prevent a SIGKILLED task to run do_exit promptly
after it has been selected for oom-killage.

Signed-off-by: Andrea Arcangeli <andrea@suse.de>

diff --git a/mm/filemap.c b/mm/filemap.c
--- a/mm/filemap.c
+++ b/mm/filemap.c
@@ -927,6 +927,16 @@ page_ok:
 		isize = i_size_read(inode);
 		end_index = (isize - 1) >> PAGE_CACHE_SHIFT;
 		if (unlikely(!isize || index > end_index)) {
+			page_cache_release(page);
+			goto out;
+		}
+
+		if (unlikely(sigismember(&current->pending.signal, SIGKILL))) {
+			/*
+			 * Must not hang almost forever in D state in
+			 * presence of sigkill and lots of ram/swap
+			 * (think during OOM).
+			 */
 			page_cache_release(page);
 			goto out;
 		}
@@ -2063,6 +2073,16 @@ static ssize_t generic_perform_write_2co
 			break;
 		}
 
+		if (unlikely(sigismember(&current->pending.signal, SIGKILL))) {
+			/*
+			 * Must not hang almost forever in D state in
+			 * presence of sigkill and lots of ram/swap
+			 * (think during OOM).
+			 */
+			status = -ENOMEM;
+			break;
+		}
+
 		page = __grab_cache_page(mapping, index);
 		if (!page) {
 			status = -ENOMEM;
@@ -2230,6 +2250,16 @@ again:
 		 */
 		if (unlikely(iov_iter_fault_in_readable(i, bytes))) {
 			status = -EFAULT;
+			break;
+		}
+
+		if (unlikely(sigismember(&current->pending.signal, SIGKILL))) {
+			status = -ENOMEM;
+			/*
+			 * Must not hang almost forever in D state in
+			 * presence of sigkill and lots of ram/swap
+			 * (think during OOM).
+			 */
 			break;
 		}
 

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

  parent reply	other threads:[~2008-01-03  2:09 UTC|newest]

Thread overview: 40+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2008-01-03  2:09 [PATCH 00 of 11] oom deadlock fixes Andrea Arcangeli
2008-01-03  2:09 ` [PATCH 01 of 11] limit shrink zone scanning Andrea Arcangeli
2008-01-07 19:11   ` Christoph Lameter
2008-01-03  2:09 ` [PATCH 02 of 11] avoid oom deadlock in nfs_create_request Andrea Arcangeli
2008-01-07 19:13   ` Christoph Lameter
2008-01-03  2:09 ` Andrea Arcangeli [this message]
2008-01-07 19:15   ` [PATCH 03 of 11] prevent oom deadlocks during read/write operations Christoph Lameter
2008-01-07 19:26     ` Andrea Arcangeli
2008-01-03  2:09 ` [PATCH 04 of 11] avoid selecting already killed tasks Andrea Arcangeli
2008-01-03  9:40   ` David Rientjes
2008-01-03 13:41     ` Andrea Arcangeli
2008-01-03 18:47       ` David Rientjes
2008-01-03 19:54         ` Andrea Arcangeli
2008-01-03 20:49           ` David Rientjes
2008-01-07 19:17   ` Christoph Lameter
2008-01-03  2:09 ` [PATCH 05 of 11] reduce the probability of an OOM livelock Andrea Arcangeli
2008-01-07 19:32   ` Christoph Lameter
2008-01-03  2:09 ` [PATCH 06 of 11] balance_pgdat doesn't return the number of pages freed Andrea Arcangeli
2008-01-07 19:33   ` Christoph Lameter
2008-01-03  2:09 ` [PATCH 07 of 11] don't depend on PF_EXITING tasks to go away Andrea Arcangeli
2008-01-03  9:52   ` David Rientjes
2008-01-03 13:29     ` Andrea Arcangeli
2008-01-03  2:09 ` [PATCH 08 of 11] stop useless vm trashing while we wait the TIF_MEMDIE task to exit Andrea Arcangeli
2008-01-03  2:09 ` [PATCH 09 of 11] oom select should only take rss into account Andrea Arcangeli
2008-01-07 19:35   ` Christoph Lameter
2008-01-03  2:09 ` [PATCH 10 of 11] limit reclaim if enough pages have been freed Andrea Arcangeli
2008-01-07 19:37   ` Christoph Lameter
2008-01-08  7:28     ` Andrea Arcangeli
2008-01-03  2:09 ` [PATCH 11 of 11] not-wait-memdie Andrea Arcangeli
2008-01-03  9:55   ` David Rientjes
2008-01-03 13:06     ` Andrea Arcangeli
2008-01-03 18:54       ` David Rientjes
2008-01-07 19:43   ` Christoph Lameter
2008-01-08  1:57     ` David Rientjes
2008-01-08  3:25       ` Nick Piggin
2008-01-08  3:37         ` David Rientjes
2008-01-08  7:42           ` Nick Piggin
2008-01-08  7:45         ` Andrea Arcangeli
2008-01-08  7:37       ` Andrea Arcangeli
2008-01-08  7:31     ` Andrea Arcangeli

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=71f1d848763c80f336f7.1199326149@v2.random \
    --to=andrea@cpushare.com \
    --cc=akpm@linux-foundation.org \
    --cc=linux-mm@kvack.org \
    --cc=rientjes@google.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox